Solar-powered oxygen delivery for the treatment of children with hypoxemia : a stepped wedge cluster randomized controlled trial

Background: Supplemental oxygen (O2) is not always available at health facilities in low- and middle-45 income countries (LMICs). Solar-powered oxygen (SPO2) delivery can overcome gaps in O2 46 access, generating O2 independent of grid electricity. Methods: In a stepped wedge cluster randomized controlled trial, SPO2 systems (photovoltaic cells, 49 battery bank, and O2 concentrator) were sequentially installed at 20 rural health facilities in 50 Uganda. Children under five years of age hospitalized with hypoxemia and respiratory signs 51 were included. The primary outcome was mortality within 48-hours of detection of 52 hypoxemia. The statistical analysis used a linear mixed effects logistic regression model 53 accounting for cluster as random effect and calendar time as fixed effect. The trial is 54 registered, ClinicalTrials.gov Identifier: NCT03851783. Findings: Between 28 June 2019 and 30 November 2021, 2405 children were recruited and included: 57 964 pre-randomisation and 1441 post-randomisation. Use of O2 increased from 484/964 58 (50%) pre-randomisation to 1426/1441 (99%) post-randomisation (p<0·0001). There were 59 104 deaths, 91 of which occurred within 48 hours of detection of hypoxaemia. The 48-hour 60 mortality was 49/964 (5·1%) in the pre-randomisation group and 42/1440 (2·9%) in the post-61 randomisation group (adjusted odds ratio 0·50, 95%CI 0·27-0·91, p=0·023). Results were 62 sensitive to alternative parametrizations of the secular trend. This represents a relative risk 63 reduction of 49% (95% CI 8·5% to 72%), and a number needed to treat with SPO2 of 45 64 (95% CI 28 to 230) to save one life. The estimated cost-effectiveness was USD$25 per 65 disability-adjusted life-year saved (95%CI 5·64 to 505). Interpretation: This pragmatic country-wide stepped wedge cluster randomized controlled trial demonstrates 68 the mortality benefit of improving O2 access with SPO2. This could serve as a model for 69 scale-up of SPO2 as one solution to O2 insecurity in LMICs.